So you want to patch turbo into the JOY1 schematic shown above. Do you want NES Max-style turbo with different buttons? Or do you want NES Advantage/TG16/SN ProPad style turbo with an on/off switch for each button's turbo feature? Either way, you'll need an oscillator set somewhere between 15 and 30 Hz to generate a signal used to modulate the output of the A, B, and Start switches before they hit the 4021 shift register.
The turbo on the Advantage was pretty nice, because the speed was adjustable. The potentiometer could be replaced with a normal resistor for a fixed speed.
For a separate turbo button, you could use the "A" and "turbo A" buttons as inputs to an AND gate (4073, 4081, 74HC08, etc.), then the output goes your 4021 to be the A button.
Also, that 120uf capacitor seems awfully high to me. I'm guessing 1uf might be more correct, or possibly 10uf at the outside.
If you don't want to have a turbo on/off switch, and would like a "normal" and "turbo" button (like on that controller) this is possible too by using an NPN transistor across the regular "A" (or B) button, with emitter grounded, and the base connected thru a 10K resistor to the oscillator via the turbo button.
As an aside, that chip in that microgenius controller is an ASIC and is not a standard part. It probably has an oscillator of some form built in, or else it just has a 2 bit or so binary counter on the load line so that it can toggle the button state every 2 or 4 or even 8 frames. That'd negate the need for an oscillator all together.
qbradq wrote:I always thought turbo controllers used a set / reset flip-flop. That could work, but you would be forced to use a 2 frame cycle.
You mean toggle them every strobe? That'd just make games freeze if they use a controller read loop that waits for the buttons to settle. A lot of games using DPCM do this to work around a bug in the NES's DMA controller.
qbradq wrote:I always thought turbo controllers used a set / reset flip-flop. That could work, but you would be forced to use a 2 frame cycle.
You mean toggle them every strobe? That'd just make games freeze if they use a controller read loop that waits for the buttons to settle. A lot of games using DPCM do this to work around a bug in the NES's DMA controller.
Seems like if you pressed B normal and B turbo at the same time (not that you'd want to, but it would be easy to accidentally), the output at pin 8 (on the '00) would be shorted to ground. That's sorta why I was thinking it should go through an AND gate, or alternatively like kevtris said, a transistor.